As soon as you have anything to do with Six Sigma, you’ll hear of DMAIC, but don’t let it put you off, it’s a lot simpler than it sounds. It stands for Define, Measure, Analyze, Improve and Control, and it’s the main framework for process improvement in Six Sigma. You simply follow the steps in order to improve the efficiency of your process.
The steps are:
- Define the aims of the project – what you want to achieve
- Measure the current system, to find where the issues are
- Analyze the data to see what the main issues are, and why they are occurring
- Improve the current system to remove the issues found
- Control the improved system so the new system is maintained
Make sure you finish each stage before starting the next (which will be confirmed by a ‘toll-gate review’), and just follow the steps one by one until you reach the end!
This is the setup; where you define what overall (broad) goals the project will have, including overall goal and schedule. It’s also where you put together a team (including training if required), work out who will be impacted by the change, and get authorization from management. You put together the Project Charter at this stage (the key output), which has a full description of the job.
- Team creation
- Choose team members
- Project Charter
- Overall aim of the project
- Get authorization from management
You go into a lot more detail at this stage, getting customers involved to get a full map of the process as it currently stands. You need to get a handle on the past to know where you are going, namely:
- What problems have been found with the current process that has led to this project?
- Has anything already been tried? If so what was the result?
- Have any potential causes already been identified?
- Collect data on issues such as yields and issues frequencies, to determine the base line of how many issues the current system is creating (this data will be analysed in more detail in the ‘Analyze’ phase next)
- Create process maps which thoroughly map how the process works, including inputs, outputs, and timings
- Closely define the current how the current process operates
- Agree on a reliable way of measuring the system
- Measure the current output / yield
Analyze is where you find what’s causing the issues (and what factors can supply improvements), and see if you can find processes elsewhere to emulate.
- See if there is a ‘best in class’ process that you can roll out to the rest of the system
- Determine the sources of any variation in the process, and the factors that lead to error
This is the good bit, where you work out your priorities for improvement, then put them into practice.
- Summarize the opportunities for improvement
- Prioritize the improvements, and choose which are to be implemented
- Define the process for the new system
- Analyze the new process for potential issues
- Implement the new system(s)
The final stage is all about not letting your hard won gains go to waste, and things slip back to their old ways.
- Measure the results of the new process to verify that the expected benefits have occurred
- Develop controls so that the gains are maintained
- Write up the project